UMTS NETWORK ARCHITECTURE | Mcgraw-Hill Education

3. UMTS NETWORK ARCHITECTURE

3.1. INTRODUCTION

This chapter provides an overview of the UMTS network architecture. Its objective is to familiarize the reader with some of the basic concepts of UMTS network, as subsequent chapters are built upon these concepts.

The chapter starts with a discussion on the basic structure of UMTS network, which divides the network into three logical parts: User Equipment (UE), Access Network (AN) and the Core Network (CN). Each of these parts is discussed in detail in the next three chapters, i.e. Chapters 4, 5, and 6 respectively.

The discussion on the logical structure of UMTS network is followed by the stratification of UMTS network into Access Stratum (AS) and Non-Access Stratum (NAS). The concept of AS and NAS is important in understanding the message flows between UE and Access Network as well as between UE and the Core Network.

The hierarchical organization of UMTS network, which divides the UMTS PLMN into Location Area, Routing Area, UTRAN Registration Area and Cells, is discussed next. Thereafter, the various addresses/identifiers used in UMTS are elaborated, upon. The next topic is service aspects and service classification in the UMTS network, with brief mention of Bearer service, Teleservice and Supplementary Service. Then, the UMTS QoS Architecture is discussed. Finally, the four QoS classes in UMTS, namely the Conversational class, Streaming class, Interactive class and Background class are elaborated upon.

3.2. BASIC STRUCTURE OF UMTS NETWORK

A typical UMTS network can be modeled on its three basic parts or sub-systems, namely User Equipment (UE), Access Network (AN) and the Core Network (CN). This basic model of the UMTS network is depicted in Figure 3.1.

Figure 3.1 Basic Structure of UMTS Network

The User Equipment (UE) is used by a subscriber/user to access the services provided by the network. To connect to the network, a UE interfaces with the Access Network using the WCDMA air interface, which is referred to as the Uu interface. Two modes of operation are used over the Uu interface: the Frequency Division Duplex (FDD) mode for the paired spectrum and the Time Division Duplex (TDD) mode for the unpaired spectrum. These modes of operation were discussed in Chapter 2.

The Access Network (AN) performs functions specific to the radio access technique. In case of UMTS, the Access Network performs functions specific to the WCDMA air interface. The Access Network has two different types of entities: the Base Transceiver Station (BTS) that terminates the radio connection with the UE, and a Base Station Controller (BSC) that controls the resources of the BTS. BSC and one or more BTS collectively form the Access Network. The BSC interfaces with the Core Network over the Iu interface.

The Core Network (CN) performs the core functions of the network, which include mobility management, call control, switching and routing. The Core Network also manages the subscription information of a subscriber and provides services based on this

information.

The basic structure of the UMTS network is similar to that used in any wireless network. In particular, it is modeled on the lines of GSM/GPRS network architecture. Thus, at the architectural level there are many similarities between the two. However, the actual protocols residing on these entities are quite different. This difference is created by the introduction of WCDMA-based air interface in the UMTS Access Network, leading to significant changes in the protocols residing at the User Equipment and the Access Network. Thus, the GSM/GPRS mobile handsets are rendered useless in a UMTS environment (unless they are backward compatible). In contrast, the Core Network of GSM/GPRS is almost entirely reused in the UMTS. Even though there are changes and enhancements in Core Network protocols, the main network entities (e.g. HLR, VLR, SGSN and GGSN) and the important Core Network protocols (e.g. MAP, GTP and ISUP) exist in the UMTS as well. This implies that upgrading the Core Network to make it compliant with UMTS standards is easier as compared to upgradation of the UE or the Access Network.

The following sub-sections provide the details of the User Equipment (UE), the Access Network (AN) and the Core Network (CN).

3.2.1. User Equipment (UE)

The User Equipment (UE) is a device used by a subscriber/user to access network services. To make its design modular, the UE is divided into two logical parts: the Mobile Equipment (ME) and the Universal Subscriber Identity Module (USIM). The logical structure of UE is depicted in Figure 3.2.

Figure 3.2 Logical Structure of User Equipment

The Mobile Equipment (ME), or the Mobile handset, is manufactured by equipment vendors. The ME is further divided into two distinct functional groups, namely, Mobile Termination (MT) and Terminal Equipment (TE). The MT performs functions like radio transmission termination, authentication, and mobility management. The TE manages the hardware (e.g. speaker, microphones, video cameras, and user display) and hosts user applications (e.g. Web browser). The division of Mobile Equipment into MT and TE is also referred to as MT-TE functionality split. As an example, a Mobile Termination (MT) unit may be physically connected to a Laptop (which acts as a TE). The same MT may also provide services to other Terminal Equipment like a camera, using a Bluetooth interface.

Besides the Mobile Equipment (containing the MT and the TE), the User Equipment also contains a Universal Subscriber Identity Module (USIM) application. The USIM contains the logic required to unambiguously and securely identify the user. In parti-cular, it contains the permanent identity of the user (called the IMSI), the shared secret key (used for authentication), phone book and a host of other information. The USIM application resides on a Smart Card that can be inserted or removed from the ME. The smart card is called the UMTS Integrated Circuit Card (UICC). The USIM on the UICC card is provided by the service provider. Hence, even if the UICC card is moved from one ME to another, the service provider and the service configuration remains the same.

Chapter 4 provides the details of the User Equipment (UE). The topics covered in the chapter include the logical structure of UE (especially the MT-TE functionality split) and details of the USIM application.

Note that in this book, the User Equipment (UE) is also referred to as Mobile Station (MS). Thus, the terms UE and MS are used interchangeably.

3.2.2. Access Network (AN)

Access Network resides between the UE and the Core Network. It performs the functions specific to the access technique. In case of UMTS, Access Network performs functions specific to the access of WCDMA air interface. The Core Network, on the other hand, may be used with any access technique. This functional split between the Core Network and the Access Network provides the flexibility to keep the Core Network fixed, while at the same time allowing for different access techniques.

The Access Network in UMTS allows two different types of access network systems to interface with the Core Network. These two systems are the Base Station Sub-system (BSS) and the Radio Network Sub-system (RNS). While BSS is the legacy of the GSM era, the RNS is the newly standardized access network for UMTS networks following Rel99 onwards 3GPP specifications. The Core Network can connect to one or both of these Access Network types.

Both types of access systems (i.e. the BSS and the RNS) have a similar structure. They comprise of a Base Station Controller (BSC) and one or more Base Transceiver Station (BTS). The BTS terminates the radio connection with the UE; the BSC controls one or more BTS. The nomenclature of BTS and BSC is specific to Base Station Sub-system (BSS). In Radio Network Sub- system (RNS), the BTS is referred to as Node B, while the BSC is referred to as Radio Network Controller (RNC). The Radio Network Sub-system is also known as the universal Terrestrial Radio Access Network (UTRAN). The Access Network comprising of BSS and RNS is depicted in Figure 3.3.

Figure 3.3 Logical Structure of Access Network

Chapter 5 provides details of the Access Network, especially the UTRAN. The topics covered in the chapter include UTRAN network architecture, the network entities (e.g. Node B and RNC), the interfaces (e.g. Iu/Iur/Iub interface), the functions performed by and the protocols used in Access Network (e.g. RRC and RANAP).

3.2.3. Core Network (CN)

In the preceding sections, the functions and structure of User Equipment (UE) and Access Network (AN) were discussed. The UE provides an interface to the end user. Thus, its functions are limited to terminating the radio interface and hosting user applications. The Access Network also provides limited functions; its scope is restricted to managing radio connection with UE and associated radio resources. Thus, apart from these two entities, there is a clear need for a sub-system that would perform the following functions:

Mobility Management: This refers to tracking the location of the UE. In a mobile network, where the position of the UE is not fixed, this is a very important function.

Call Control: This refers to establishment and release of voice call between the UE and an end-point. Here, the end-point may be another UE or even a point outside the mobile network (e.g. a fixed telephone of a PSTN). Whatever be the case, a call control function is required to establish/release a voice connection between UE and an end-point.

Switching: This refers to switching a voice call between the UE and an end-point. The switching function is performed after a voice connection is established.

Session Management: This refers to establishment and releases of sessions for data transfer between UE and an end- point. The end-point is typically outside the mobile network (e.g. an Internet Server).

Routing: This refers to routing of data packets between the UE and an end-point.

Authentication: This esures that the user availing the service is authenticated.

Equipment Identification: This ensures that the handset through which services are availed is genuine (and not stolen).

The above are just some of the functions that are performed by the Core Network (CN). In simple terms, the Core Network consists of the entities that provide support for various network features and services and performs functions like mobility management, call control, switching, session management, routing, authentication and equipment identification.

It is evident from above that there are two classes of traffic handled by the Core Network, namely,v oice and data. The 2G mobile networks, like GSM networks, were designed primarily for voice. The GPRS networks provided capability for data transfer. Based on the fact that the UMTS Core Network is an evolved GSM/GPRS core network, the former is divided into two domains: the Circuit Switched (CS) domain and the Packet Switched (PS) domain. The CS domain provides services related to voice transfer, the PS domain to those related to data transfer. The entities of the Core Network and its decomposition into the CS and PS domain is depicted in Figure 3.4.

Figure 3.4 Logical Structure of Core Network

The CS domain uses Circuit-Switched (CS) connections for communication between UE and the destination. A CS connection is defined as a connection for which dedicated network resources are allocated at the time the connection is established and are freed when the connection is released. An example of CS connection is the connection established in the PSTN network during a telephonic conversation.

To establish/release CS connections and to switch voice streams, a switching entity is required. For this, the CS domain has the Mobile Switching Center (MSC). Alongside the MSC, there is another entity in the CS domain, called the Visitor Location Register (VLR). The VLR contains the subscriber profile obtained from the Home Location Register (HLR). MSC queries the VLR for subscriber information and provides services to the subscriber based on the queried information. It is customary to represent

MSC and VLR as one entity: MSC/VLR. Apart from MSC/VLR, the CS domain has the Gateway Mobile Switching Center (GMSC). The GMSC provides connectivity to external CS networks (including the CS domain of other UMTS networks and the PSTN networks).

The PS domain uses Packet-Switched (PS) connections for communication between UE and the destination. A PS connection is defined as a connection that transports the user information using autonomous concatenation of bits called packets; each packet is routed independently from the previous one. An important aspect of PS connection is that resources are not reserved for a connection; rather, they are shared between various communicating entities. This sharing of resources results in better resource utilization. An example of PS connection is the connectionless transfer of IP datagrams in the Internet.

To route packets in the PS domain, a routing entity is required. For this, the PS domain has the Serving GPRS Support Node (SGSN). Unlike in the CS domain, where one entity holds the database (VLR) and another switches CS the connections (MSC), in PS domain, the SGSN performs both the functions. Apart from SGSN, the PS domain has the Gateway GPRS Support Node (GGSN) which performs functions similar to those of GMSC (i.e. GGSN provides connectivity to external PS networks).

Apart from entities belonging to the CS and PS domain, there are entities that are common to both the domains. Important among these is the Home Location Register (HLR) that is located in the subscriber's home network. The HLR holds the permanent and subscribed information of the subscriber. The permanent information includes the permanent identity of the user (called the IMSI). The subscribed information includes information about the services that are provisioned in the HLR, based on the services subscribed to by the subscriber.

Then, there is the Authentication Center (AuC), which holds authentication information. This information is used for authentication and other security-related functions. It is customary to represent the AuC as a part of HLR. Thus, the term AuC/HLR is used to represent the entity that performs the functions of HLR and AuC.

The entities common to CS and PS domain also include Equipment Identity Register (EIR). The EIR monitors the legitimacy of a User Equipment (UE) used in the UMTS network.

Apart from HLR, AuC and EIR, there are few other common entities like SMS Gateway MSC (SMS-GMSC) and SMS Interworking MSC (SMS-IWMSC). There are also service specific entities like Gateway Mobile Location Center (GMLC), Camel entities and Cell Broadcast Center (CBC). For sake of simplicity, these entities are not depicted in Figure 3.4.

Chapter 6 provides details of the Core Network. The topics covered include Core Network architecture, its entities (including the common entities, the CS and PS domain entities and service-specific entities), interfaces (for both CS and PS domain), the functions performed by Core Network and the protocols used in it (e.g. MAP, GTP and ISUP).

Apart from the CS and PS domain, another sub-system, called the IP Multimedia Sub-system (IMS), is introduced in Rel5 specifications. The IMS uses the services of PS domain for providing IP based multimedia services. To avoid complication, the IMS too is not depicted in Figure 3.4. Chapter 16 provides the details of this sub-system.

3.2.3.1. Domain Split in Core Network

The CS and PS domain divide the Core Network on the basis of its functionality. Another way to divided CN is based on its position with respect to the user. This classification divides the Core Network into Serving Network Domain, Home Network Domain and Transit Network Domain (see Figure 3.5).

Figure 3.5 Domain Split in Core Network

The serving network domain is defined as that part of the Core Network which is connected to the access network currently providing access to a user. Thus, the serving network is defined in the context of a particular user. It is responsible for switching and routing calls/packets (i.e. transfer of user information from source to destination), for which it interacts with the home network to obtain subscriber information. It is possible that the serving network is the home network, in which case it already has the subscriber information. However, it is clear that the serving network need not necessarily be the home network. The serving network changes with the change in location of the user.

The home network, in contrast, performs functions independent of the location of the user. The home network contains the permanent data of the user (e.g. permanent subscriber identity) and subscriber information. Thus, it is responsible for the management of subscription information of the user.

The transit network domain is an optional part in communication between source and destination. It is required when the destination party is outside the serving net-work. Thus, the transit network lies between the serving network and the destination network.

3.3. ACCESS STRATUM AND NON-ACCESS STRATUM

One way of modeling the UMTS network is to divide it into UE, Access Network and Core Network. A different way of modeling it is by dividing it into Access Stratum (AS) and Non-Access Stratum (NAS). The AS protocols provide the means to carry information over the air interface as well as to manage the resources of the air interface. In contrast, the NAS protocols are those that apply between UE and the Core Network, for which the access stratum acts as a relay (see Figure 3.6).

Figure 3.6 Access Stratum and Non-Access Stratum

3.3.1. Access Stratum (AS)

The Access Stratum (AS) provides the means to carry information over the air interface and also to manage the resources of the air interface. It contains parts of the UE and parts of the Access and Core Network.

The Access Stratum is further divided into two distinct components: the Uu stratum and the Iu stratum.

The AS uses the Uu stratum for communication between the UE and the Access Network. The Uu stratum is used to manage the radio resources between the UE and the Access Network. The Uu stratum protocols include Medium Access Control (MAC), Radio Link Control (RLC), Broadcast/Multicast Control (BMC), Packet Data Convergence Protocol (PDCP) and Radio Resource Control (RRC) protocol. Among these, the RRC is the main signaling protocol between UE and Access Network (in particular, between the UE and the RNC).

Chapter 5 provides the details of the Uu stratum protocols and Chapter 7 the details of the Uu stratum procedures (specifically, the RRC procedures).

The AS uses the Iu stratum (AN-CN interface) for communication between the Access Network and the Core Network. The Iu stratum is used by the Core Network to manage the resources provided by the Access Network to the UE. The Radio Access Network Application Part (RANAP) is the main Iu stratum protocol used between RNC and MSC/VLR and between RNC and SGSN. Chapter 5 provides details of the RANAP protocol and Chapter 8 the details of the Iu stratum procedures.

In simple terms, the Access Stratum (AS) provides services to the Non-Access Stratum (NAS). One of the important services provided by AS is to transport NAS messages between NAS entities.

The protocols in Access Stratum and Non-Access Stratum are depicted inF igure 3.7. The boxes shaded in gray are the Uu stratum protocols. It is customary to refer to only the Uu stratum protocols as Access Stratum protocols. The Iu stratum protocols are seldom referred to as part of Access Stratum protocols. This can best be explained if the AS and NAS protocols are viewed from the UE point of view. For a UE, the Iu inter-face is not visible because it can be replaced by a relay function that delivers all messages received by the AS layer directly to the NAS layer protocols. Given this, the protocols between UE and Access Network (over Uu interface) form part of the AS protocols and those between the UE and the Core Network form part of the NAS protocols.

Figure 3.7 Protocols in Access Stratum and Non-Access Stratum

3.3.2. Non-Access Stratum (NAS)

The Non-Access Stratum (NAS) protocols are those that apply between UE and the Core Network. For these protocols, the access stratum (i.e. the UTRAN) acts as a carrier/transport. The NAS protocols are not terminated at the UTRAN (they are terminated at the Core Network).

The NAS protocols are depicted in Figure 3.7. There is a Mobility Management (MM) layer for CS domain and GPRS Mobility Management (GMM) layer for PS domain. The MM/GMM procedures enable mobility of user terminals, such as keeping track of the subsciber's present location. During mobility management, it is also ensured that the identity of the user is kept confidential and that only authenticated users can avail network services. Chapter 9 provides details of the Mobility Manage-ment (MM) and GPRS Mobility Management (GMM) procedures.

A number of protocols reside over the MM/GMM layer. This includes the Call Control (CC), Session Management (SM), Supplementary Service (SS) and Short Message Service (SMS) protocols.

The Call Control procedures involve handling of the mobile-originated (MO) and mobile-terminated (MT) calls. These procedures, also referred to as Call Handling (CH) procedures, are detailed in Chapter 10.

In the PS domain, there is no concept of calls. Hence, the call handling procedures are not applicable here. However, there is an analogous concept, which is termed as sessions. A session can be viewed as a context maintained by the UE and the SGSN/SGSN for information exchange in the PS domain. Chapter 11 provides details of the Session Management (SM) procedures.

Supplementary Services (SS) are modification or supplement to the basic services. Examples of SS are Call Forwarding and Call Barring. Chapter 12 provides details of the SS procedures.

Short Message Service (SMS) is the means by which a short text (of up to 160 characters) can be exchanged between the UE and a Short Message Service Center (SMSC). Chapter 13 provides details of the SMS procedures.

3.4. HIERARCHICAL NETWORK ORGANIZATION

To support mobility of a subscriber from one location to another, the UMTS network architecture is organized as a multi-tier hierarchical structure. This hierarchical structure enables a particular network entity to have only that much information as is required for its functioning. For example, a VLR has information only of the Location Area of an MS but does not know the exact cell location. Despite this, the exact cell location can be determined through paging when required. The hierarchical division is done to reduce the storage/processing load on the network entities (like the VLR), save radio resources and battery consumption of the MS.

The hierarchical structure of UMTS network is depicted in Figure 3.8. As shown in the figure, at the lowest level of UMTS hierarchy is the cell. At the next level is the UTRAN Registration Area (URA), which is a collection of cells. Then comes the Location Area (LA) and the Routing Area (RA). A collection of one or more cells forms a Location Area (LA) for the CS domain, and a Routing Area (RA) for the PS domain. Further, a Location Area contains one or more Routing Areas and a Routing Area contains one or more UTRAN Registration Areas. At the highest level of the hierarchy is a Public Land Mobile Network (PLMN), which is not depicted in the figure.

Figure 3.8 Hierarchical Structure of UMTS Network

The hierarchical structure of UMTS network is explained in greater detail in the following sub-sections.

3.4.1. Public Land Mobile Network (PLMN)

In UMTS, at the highest level of the hierarchy is a Public Land Mobile Network (PLMN). A PLMN is defined as a telecommunications network providing mobile cellular services. A PLMN is uniquely identified by its PLMN identifier.

The PLMN identifier comprises of Mobile Country Code (MCC) and Mobile Network Code (MNC), as shown in Figure 3.9. The MCC is of three digits and identifies the country to which the Public Land Mobile Network (PLMN) belongs. The next two or three digits of the PLMN identifier are the Mobile Network Code (MNC). The MNC identifies a particular PLMN within a country. It is recommended that within a country identified by the MCC, all PLMN either use only two or else three digits for MNC. A mixture of the two schemes is not recommended.

Figure 3.9 Structure of PLMN Identifier

3.4.2. Location Area (LA)

Location Area (LA) is defined as an area in which an MS may move freely without updating its current location at the VLR. In case an MS moves outside its location area, it informs the VLR of its current location through the location update procedure.

A location area includes one or more cells. The reason for grouping of cells into location area is to facilitate efficient location

management. To understand this, note that location management requires tracking the current location of the MS so that a terminating call can be delivered to the MS. Since the MS updates its location only at the change of location area, the VLR has accurate information on this. When a terminating call for an MS arrives, the VLR pages the MS to seek the exact location of the MS (in terms of its current cell location). Upon receiving the paging request, the MS responds with information on its current cell location. This information is used to set up a connection with the MS.

Hence, it is evident that if the location area is as small as a cell, there is no need to page the MS. However, this would require constant activity between the MS and VLR whenever the MS moves to a new cell, resulting in consumption of radio resources and battery power. If the location area is very large, the paging has to be performed in a very large area, which is undesirable. Thus, grouping of cells into location area allows us to achieve a balance between the accuracy of information maintained at the VLR as against the uplink radio capacity and the battery power consumed in the process.

Each location area is uniquely identified by a Location Area Identity (LAI). The structure of LAI is shown inF igure 3.10. The MCC and MNC are the same as Mobile Country Code and Mobile Network Code of the PLMN to which the LA belongs. The last two octets of the LAI are the Location Area Code (LAC) that identifies a location with a PLMN. Collectively, the LAI forms a unique identifier for a location area across all PLMNs.

Figure 3.10 Structure of Location Area Identity (LAI)

3.4.3. Routing Area (RA)

The Routing Area (RA) for PS domain is analogous to the location area for the CS domain. Routing area is defined as an area in which an MS may move freely without updating its current location at the SGSN. In case an MS moves outside its routing area, it informs the SGSN of its current location through the routing area update procedure.

Like a location area, a routing area too may include one or more cells. Grouping of cells into a Routing Area facilitates efficient location management, like in the case of a location area, where balance is achieved between the frequency of location updates and the area in which paging is done for mobile-terminated sessions.

One important difference between a routing area and location area is that the former is always contained within a location area. In other words, a location area may contain one or more routing areas.

Each routing area is uniquely identified by a Routing Area Identity (RAI). Since a routing area is a subset of a location area, the Routing Area Identity (RAI) is derived from Location Area Identity (LAI). In fact, the RAI is LAI plus a Routing Area Code (RAC) of 1-octet. The RAC uniquely identifies a routing area in a location area. Putting it simply, RAI= LAI + RAC.

3.4.4. UTRAN Registration Area (URA)

A UTRAN Registration Area (URA) is defined as an area covered by a number of cells. It is only internally known in the UTRAN. The URA is used to provide a layer of abstraction between cells and the routing area. A URA contains one or more cells and a routing area contains one or more URA. The URA is used to track the location of an MS within the UTRAN. The use of URA for mobility management is explained in Chapter 9. A URA is uniquely identified using the URA identity.

3.4.5. Cell Global Identity (CGI)

At the lowest level of UMTS hierarchy is the cell. Each cell is identified by the Cell Identity (CI). A CI is unique within a location area. To identify a cell uniquely across PLMNs, an identity called the Cell Global Identity (CGI) is defined. CGI is obtained by the concatenation of Location Area Identity and the Cell Identity. The structure of CGI is depicted in Figure 3.11.

Figure 3.11 Structure of Cell Global Identity (CGI)

3.5. ADDRESSES AND IDENTIFIERS

In UMTS, a number of identifiers are used for the purpose of addressing and identification. Each identifier serves a specific purpose. First comes the International Mobile Subscriber Identity (IMSI) that uniquely identifies a subscriber. An IMSI may be associated with multiple Mobile Subscriber ISDN (MSISDN) numbers. The MSISDN can be viewed as the mobile phone numbers or the service identity. Apart from these two identifiers, there is a temporary identifier, TMSI, which is used to hide the IMSI. There are other temporary identities as well, the need and functions of which are detailed later in this section.

While MSISDN is relevant in the CS domain, the equivalent identity in the PS domain is the Packet Data Protocol (PDP) address. In simple terms, the PDP address identifies the network address using which entities outside the PS domain communicate with the MS.

Apart from these, there is the International Mobile Equipment Identity (IMEI), which uniquely identifies a MS.

Then there are the E.164 addresses, used to identify network entities.

All these identifiers and addresses are explained in the following sub-sections (also see Table 3.1). The reader is referred to 3GPP TS 23.003 for complete information on numbering, addressing and identification schemes used in the UMTS network.

Table 3.1 UMTS Addresses and Identifiers

Identity Description Composition

IMSI Permanent identity that uniquely identifies a subscriber. MCC + MNC + MSIN

MSISDN Service identity that is used for communication with a subscriber. CC + NDC+ SN

TMSI Temporary identity that is used to hide the permanent identity IMSI of a subscriber. Four octets (chosen by operator)

LMSI Temporary identity that is used by VLR to optimize database search. Four octets (allocated by VLR)

MSRN Temporary identity that is allocated by VLR and is used to route calls directed to a MS. CC + NDC+ SN

RNTI Temporary identity used as UE identifiers to exchange signalling messages between UE and Refer 3GPP TS 25.401 UTRAN.

PDP Address Static or dynamic network address used to communicate with other entities of a Packet Data Typically IPv4 or IPv6 Network (PDN). address

IMEI Permanent identity that uniquely identifies an MS. TAC + SNR

Location Refers to the geographical position of the MS in terms of standardized co-ordinates. CC + NDC+ LSP Number

E.164 address Used by MSC, GMSC, SGSN, GGSN, EIR, HLR and VLR for the purpose of signaling. CC + NDC+ SN

GSN address Used by the GSNs to communicate with each other over IP backbone. IPv4 or IPv6 address

RNC identifier Used to uniquely identify an RNC. MCC + MNC + RNC-id

3.5.1. Subscriber Identity

A subscriber is uniquely identified by its International Mobile Subscriber Identity (IMSI). The IMSI is stored in the USIM and kept hidden from ordinary access. As shown in Figure 3.12, the IMSI is divided into three distinct parts. The first three digits of the IMSI form the Mobile Country Code (MCC). The MCC identifies the country of domicile of the mobile subscriber. The next two or three digits form the Mobile Network Code (MNC). The MNC identifies the home PLMN of the subscriber. The last field of IMSI is the Mobile Subscriber Identification Number (MSIN). The MSIN uniquely identifies a subscriber within a PLMN. The combination of MNC and MSIN is called the National Mobile Subscriber Identity (NMSI).

Figure 3.12 Structure of IMSI

3.5.2. Service Identity

The mobile number used to contact a person is the Mobile Subscriber ISDN (MSISDN) number and not the IMSI. Thus, an MSISDN can be viewed as a service identity because a subscriber may have multiple MSISDN, where each MSISDN identifies a particular service.

The MSISDN numbers are based on the ISDN numbering plan and are allocated in such a manner that fixed line ISDN or PSTN subscribers can call any mobile subscriber. The ISDN numbering plan is based on ITU-T specification E.164.

Figure 3.13 shows the structure of MSISDN. Like IMSI, an MSISDN number is made up of three distinct parts: a Country Code (CC), a National Destination Code (NDC) and a Subscriber Number (SN). There is a one-to-one analogy between the elements of IMSI and those of MSISDN. The basic difference between the two is the number of digits allocated to individual elements. The CC is from one to three digits.

Figure 3.13 Structure of MSISDN

The MSISDN can have a maximum of 15 digits. The size of National (Significant) Number depends upon the size of the CC and can be of a maximum of 14 digits (when CC is of one digit).

3.5.3. Temporary Identities

Apart from IMSI and MSISDN, there are temporary identifiers used for specific purposes. These temporary identifiers are as follows:

Temporary Mobile Subscriber Identity (TMSI)

Local Mobile Station Identity (LMSI)

Mobile Station Roaming Number (MSRN)

Radio Network Temporary Identity (RNTI)

3.5.3.1. Temporary Mobile Subscriber Identity (TMSI)

From security point of view, there is a requirement to hide the permanent identity IMSI of the subscriber. For this, a temporary identity TMSI (and not IMSI) is used on the air interface. The TMSI, or Temporary Mobile Subscriber Identity, is allocated by the VLR or SGSN. It is also possible that two temporary identities are used, one for CS and another for PS domain. Under such circumstances, the identities are called TMSI and P-TMSI for CS and PS domain respectively. The TMSI has only local significance and is applicable within the area controlled by VLR (or SGSN).

The TMSI consists of four octets. The exact encoding of TMSI is chosen by agreement between the network operator and equipment manufacturer to suit local needs.

3.5.3.2. Local Mobile Station Identity (LMSI)

For the purpose of optimizing database search, a VLR may use a local identifier called the Local Mobile Station Identity (LMSI).

The VLR sends the LMSI to the HLR during message exchange along with IMSI/MSISDN. The HLR does not use the LMSI but keeps it along with IMSI/MSISDN in its database. In all further correspondence with the VLR, the HLR includes the LMSI sent earlier by the VLR. The VLR then uses the LMSI to optimize database search.

The LMSI consists of four octets and is allocated by the VLR.

3.5.3.3. Mobile Station Roaming Number (MSRN)

To facilitate roaming, a VLR allocates a roaming number called the Mobile Station Roaming Number (MSRN). The MSRN is used to route calls directed to a MS. When a mobile-terminated call is received by GMSC, it queries the VLR (via HLR) for a number, using which it can route the call. The VLR allocates a MSRN for the MS and passes it to HLR, which in turn forwards it to GMSC. The GMSC then uses the MSRN to route the call to the MS via MSC/VLR.

The MSRN is of the same format as the MSISDN, but is not the same as the MSISDN. The MSRN is allocated by the visited network according to the numbering plan of the visited PLMN. In certain cases, the MSRN may be the same as the MSISDN (for example, when the subscriber is in the home network).

3.5.3.4. Radio Network Temporary Identity (RNTI)

While TMSI, LMSI and MSRN are allocated by Core Network, there are temporary identities allocated by the UTRAN. One such temporary identity is the Radio Network Temporary Identity (RNTI). The RNTI is used as a UE identifier to exchange signalling messages between UE and UTRAN. There are various types of RNTI, out of which only the s-RNTI is discussed here.

The s-RNTI is allocated by the Serving RNC, which is in charge of the radio connections between the UE and UTRAN. The Serving RNC allocates the s-RNTI for those UE that have a RRC connection. The s-RNTI is used by the UE to identify itself to the Serving RNC and is used by the Serving RNC, which in turn uses it (s-RNTI) to address a particular UE.

Apart from s-RNTI, there are d-RNTI, c-RNTI, u-RNTI and few other identifiers. For details of these, the reader is referred to 3GPP TS 25.401.

3.5.4. PDP Address

For an MS to communicate with entities of a Packet Data Network (PDN), it must have an address applicable in the PDN. Note that the PDN lies outside the PLMN, which implies that the addresses in the PLMN (like IMSI) are alone not sufficient for communication with PDN entities. Since the most common PDN is based on the Internet Protocol (IP), a MS must have an IP address for communicating with other entities in the PDN. The IP address may be an IPv4 or an IPv6 address. In either case, the MS must have an address, called the Packet Data Protocol (PDP) address, to communicate with entities in a PDN.

The PDP address is assigned either statically or allocated dynamically by the GGSN. A static PDP address is allocated by the network operator of the home PLMN. Since the allocation is static, it is of permanent nature.

However, network addresses are a scarce resource and it does not make sense to allocate them on a permanent basis, more so because a subscriber may not need to use one all the time. Hence, the addresses are generally allocated dynamically, so that a small set of these may be shared between a large number of subscribers.

A dynamic PDP address is allocated during the activation of PDP context. A PDP context can be viewed as a set of information maintained by UE, SGSN and GGSN. It contains a PDP type (that identifies the type of PDN, for example IPv4); the PDP address (say a dynamically allocated IPv4 address); QoS information; and other session information. Activating a PDP context refers to creating the PDP context at the UE, SGSN and GGSN so that the UE can communicate with an entity in PDN using the PDP address maintained in the PDP context. After the communication is over, the PDP context is deactivated.

Chapter 11 provides details of PDP context and also the procedures related to its activation and deactivation.

3.5.5. Equipment Identity

A MS is identified by its International Mobile Equipment Identity (IMEI). The IMEI is a 15-digit identifier (its structure is shown in Figure 3.14). The first eight-digits are the Type Allocation Code (TAC). The next six-digits from the Serial Number (SNR). The last digit is spare and is set at 0.

Figure 3.14 Structure of IMEI

The IMEI is used to track stolen a MS. The IMEI of a handset can be known by typing the string* #06# (star hash 0 6 hash) on the MS.

3.5.6. Location Number

In section 3.4, the location of a subscriber was defined in terms of Location Area, Routing Area, UTRAN Registration Area and Cells. This location referred to the location of the MS in the hierarchical structure of the network. There is yet another notion of location (or position) of the MS. According to this notion, the location (or position) refers to the geographical position of the MS in terms of standardized co-ordinates. This information is used by application service providers to provide specialized services, also referred to as Location Services (LCS). An example of LCS is providing the list of restaurants around a given geographical location to the MS. Chapter 13 gives the details of Location Services (LCS).

To provide Location Services, a location number is required. The location number defines a specific location within a PLMN. It contains the Country Code (CC), National Destination Code (NDC) and a Locally Significant Part (LSP). The structure of the location number is depicted in Figure 3.15. The exact structure of LSP is a matter of agreement between the PLMN operator and the national numbering authority in the country containing the PLMN.

Figure 3.15 Structure of Location Number

3.5.7. Identifying Network Entities

The preceding sections elaborated upon the identifiers used for subscriber identification, service identification, and equipment identification. Apart from these, there are identifiers used for addressing network entities.

The core network entities, including MSC, GMSC, SGSN, GGSN, EIR, HLR and VLR, are identified using the E.164 numbers. Recall from Section 3.5.2 that the MSISDN number is based on E.164 format.

Apart from the E.164, the SGSN and GGSN also require a GSN address (i.e. an IP address). This is because the GSNs communicate with each other using the IP protocol. The structure of the GSN address is depicted in Figure 3.16. The format of the address depends upon whether the protocol used is IPv4 or IPv6. Depending upon this, the GSN address part is of 4 bytes of 16 bytes.

Figure 3.16 Structure of GSN Address

In the UTRAN, the RNC is uniquely identified by the RNC identifier. The globally unique RNC identifier comprises of the PLMN identifier (MCC + MNC) and a RNC identifier.

3.6. SERVICE ASPECTS

One of the important developments in standardization of services in UMTS is the realization that instead of standardizing the services, it is much better to standardize service capabilities. In 2G and 2.5G networks, complete sets of teleservices, applications and supplementary services have been standardized. The net result of this is that a new service or changes in existing service requires considerable effort. Further, standardization of services prevents operators from offering distinct or specialized services. In such an environment, the time it takes for the service to become commercially available is also quite long.

To change this trend, instead of standardizing the services, the service capabilities are standardized. For this, two things are required. The first requirement is the presence of a wide variety of bearers with their specified QoS parameters. These bearers help in providing lower layer capabilities to satisfy the QoS requirements of a wide variety of services. This is different from providing standardized services along with their standardized bearer capabilities. In the first case, technological improvements in service realization (e.g. a better codec for voice transfer) would result in use of a bearer with lower bandwidth. In the second case, however, as the service is tied to the bearer, it is very difficult to change the applications or the bearers. In simple terms, the idea is to decouple a service from its underlying bearer and to provide a set of bearers so that the QoS requirements of a wide variety of services can be satisfied.

The second requirement is to have the necessary mechanisms in place to realize a service. For this, certain aspects like the functionality provided by various network elements, the communication between these elements and the data stored in them must be defined. This implies that an application developer must have standardized mechanisms (which is different from a standardized service) to provide different applications to the subscribers.

As an example, the IP Multimedia Subsystem (IMS) described in Chapter 16 primarily talks about a framework for realizing the services. Thus, instead of standardizing a host of teleservices or supplementary services, a set of bare minimum services is standardized. For the rest, a framework is provided so that an array of services can be provided through Applications Servers. However, there is no indication of what these services are; only the means to provide them have been defined.

3.7. SERVICE CLASSIFICATION

Services in UMTS (Figure 3.17) are classified into various categories, which are as follows:

Bearer Services (includes circuit bearer services and GPRS-based packet bearer services)

Other Bearer Service (e.g. SMS)

Teleservices

Supplementary Services

Value Added Service

IP Multimedia Service

Figure 3.17 Service Classification in UMTS Network

For details of above services, the reader is referred to 3GPP TS 22.101.

3.7.1. Bearer Services

Bearer Services refer to services that provide the capability of transmission of signals between two communicating entities. As the name suggests, the bearer service defines the lower layer capabilities as seen in the context of OSI reference model used in communication. Thus, the bearer service provides a communication link between two entities for information transport. There is the freedom to use any higher layer protocol over the bearer services.

A particular bearer service is defined by a set of characteristics that distinguishes it from other bearer service. Each characteristic of the set has a particular value such that the collection of the values of a particular bearer service uniquely defines the service.

The different characteristics associated with a bearer service are broadly classified asi nformation transfer and information quality. The parameters defining the nature of information transfer are as follows:

Nature of Service: This refers to whether the service is connection-oriented or connectionless.

Traffic Type: It refers to the basic nature of traffic. The different types of traffic are:

– Guaranteed (Constant Bit Rate)

– Non-Guaranteed (Dynamically Variable Bit Rate)

– Real-Time Dynamically Variable Bit Rate with minimum guaranteed bit rate

Traffic Characteristics: These indicate whether the service is point-to-point or point-to-multipoint. Point-to-point traffic is further classified into uni-directional, bi-directional symmetric and bi-directional asymmetric. Point-to-multipoint traffic is

further classified as multicast and broadcast. Note that point-to-multipoint by nature is uni-directional.

The parameters defining the nature of information quality are as follows:

Minimum Transfer Delay: This refers to the time taken to transfer the information from one access point to its delivery at the other access point. Minimum transfer delays are important in time-sensitive applications (e.g., voice conversation and video conferencing) because after inordinate delay, data becomes useless for that application. For example, if a movie frame reaches its destination after its succeeding frames have been viewed, it is of no use and has to be discarded.

Delay Variation: This parameter refers to variation in delay over time. The transient changes in network load cause jittery transmission, affecting real-time services.

Bit Error Ratio: This is the ratio of undetected bit error versus the total transferred information bits. While some data loss is permissible in voice conversation, it is strictly prohibited in data applications.

Data Rate: This has to do with the amount of data transferred between the two access points in a given period of time. The typical rates supported in UMTS are 384Kbps in outdoors and up to 2Mbps indoors.

Specific values of aforementioned characteristics define a particular bearer service. There are two basic categories of bearer services, namely Circuit Bearer Services and Packet Bearer Services. The Circuit bearer services are defined in 3GPP TS 22.002. The services include synchronous bearer service and asynchronous bearer service. While the syn-chronous mode supports the Transparent (T) data service, the asynchronous mode supports both Transparent (T) data service and Non-Transparent (NT) data service.

The Packet bearer services (also referred to as GPRS bearer services) are defined in 3GPP TS 22.060. These services provide means to transmit data between user-network access points and are primarily used to carry IP packets.

3.7.2. Other Bearer Service

As mentioned in Section 3.7.1, there are two types of bearer services, namely the circuit bearer services and packet bearer services. Apart from these bearers, there are other services that can be used as bearer services. These include Short Message Service (SMS), Unstructured Supplementary Service Data (USSD) and User-to-User Signaling (UUS).

The SMS is actually a Teleservice (as explained in Section 3.7.3). However, SMS can be used as a bearer over which applications can be built (e.g. a dating application). The details of Short Message Services are given in Chapter 13.

The Unstructured Supplementary Service Data (USSD) provides a mechanism whereby mobile users and PLMN operators can communicate with each other using means that are transparent to the MS and the intermediate network entities. This mechanism allows the development of services that are operator-specific. The Unstructured Supplementary Service Data (USSD) service is explained in Chapter 12.

The User-to-User Signaling (UUS) is actually a Supplementary Service but can act as a bearer. The UUS allows a subscriber to send (or receive) a limited amount of subscriber generated information to (or from) another user in the call. This information is passed transparently through the network without any modification of the contents. The network does not try to interpret or act upon this information. The User-to-User Signaling (UUS) service is also explained in Chapter 12.

3.7.3. Teleservices

Teleservices are defined as services that provide the full capabilities for communication by means of terminal equipment and network functions. In simple terms, the scope of a Teleservice is not restricted to merely providing transport of user information, like the bearer service. A Teleservice provides complete services as seen from user's point of view.

A bearer capability is associated with every Teleservice, and it defines the technical characteristics of a Teleservice. Here, bearer capability merely refers to the lower layer capabilities (i.e. bearer capabilities) of the Teleservice. Figure 3.18 depicts the difference between Bearer service and Teleservice. As shown in the figure, while the former extends till the mobile termination, the latter extends till the terminal equipment, providing complete services to the user.

Figure 3.18 Basic Telecommunication Services

The different groups of Teleservices, along with the individual Teleservice as defined in 3GPP TS 22.003, are as follows:

Speech Service: It includes plain speech service and emergency calls.

Short Message Service: It includes mobile-originated and mobile-terminated Short Message Service (SMS). It also includes the Cell Broadcast Service (CBS).

Facsimile Transmission: It includes alternate speech and facsimile group 3 and automatic facsimile group 3.

Voice Group Service: It includes Voice Group Call Service (VGCS) and Voice Broadcast Service (VBS). The VGCS enables a calling subscriber to establish a Voice Group Call to destination subscribers belonging to a predefined Group Call Area and Group identity. The Group Call Area and Group identity collectively identify a Voice Group Call. On similar lines, VBS enables a calling subscriber to distribute speech into a predefined geographical area to reach all or a group of service subscribers located in this area.

3.7.4. Supplementary Services

Supplementary services are those that modify or supplement the basic services (i.e. bearer service and teleservice). Unlike the basic services that are independent in themselves, supplementary services do not have any independent existence. Examples of supplementary services are Call Forwarding and Call Barring. A network operator may or may not offer supplementary services to the subscribers. Supplementary services are explained in detail in Chapter 12.

3.7.5. Other Services

Apart from Bearer and Teleservice, there are value-added non-call related services like email, MMS and WWW. These services can run over various types of bearers.

Then there are the SIP based IP Multimedia Services that use GPRS as bearers. Chapter 16 provides details of IP Multimedia Services.

3.7.6. Toolkits

In order to create or modify the various services, there are standardized toolkits available by 3GPP such as CAMEL, Location Service (LCS) or external solutions (e.g. Internet mechanisms). Pre-paid is an example of an application created with toolkits that may apply to all of these services categories. Chapter 13 provides details of CAMEL and LCS.

3.8. QUALITY OF SERVICE (QoS) ARCHITECTURE

The Quality of Service (QoS) Architecture in UMTS defines how various entities interact (at various levels of abstraction) to provide end-to-end QoS. The QoS Architecture in UMTS is depicted in Figure 3.19.

Figure 3.19 QoS Architecture in UMTS Network

To realize a certain QoS, a Bearer Service with clearly defined characteristics and functionality is set up from the source to the destination of the service. This bearer service includes all aspects necessary to provision and get a QoS as described by a contract. These aspects include the control signaling, user plane transport and QoS management functionality.

At the highest level, the end-to-end QoS applies between two Terminal Equipments (TEs). The scope of UMTS, however, is limited to providing QoS guarantees from the Mobile Termination (MT) till the edge of the Core Network. The guarantees cease to apply outside the PLMN (corresponding to external bearer), as what happens outside cannot be ascertained by the UMTS network. Similarly, the local bearer service between the MT and the TE is also outside the scope of UMTS bearer service. This is because the interface between MT and TE is not standardized. Nor is it necessary for such a distinction to always exist.

The UMTS bearer is provided by using the Radio Access Bearer (RAB) Service and the Core Network Bearer Service. The RAB is defined as the service that the access stratum provides to the non-access stratum for transfer of user data between MT and CN. As depicted in Figure 3.19, the RAB Service extends between the MT and the edge of the Core Network. It provides confidential transport of signaling data and user data between MT and CN Edge Node with the QoS necessary for the negotiated UMTS Bearer Service or with the default QoS for signaling. This service is based on the characteristics of the radio interface and is maintained for a moving MT. The RAB Service itself is realized using Radio Bearer Service and the Iu Bearer Service.

The role of the Radio Bearer Service is to cover all aspects of the radio interface transport. The Radio Bearer is defined ast he service provided by the Layer 2 of the access stratum for transfer of user data between MT and UTRAN. The Radio Bearer Service uses the UTRA FDD/TDD.

The Iu-Bearer Service, together with the Physical Bearer Service, provides the transport between UTRAN and CN. Iu bearer services for packet traffic provide different bearer services for a variety of QoS.

The Core Network Bearer Service of the UMTS core network connects the UMTS CN Edge Node with the CN Gateway to the external network. The role of this service is to efficiently control and utilize the backbone network in order to provide the contracted UMTS bearer service. The UMTS packet core network supports different backbone bearer services for a variety of QoS.

3.9. UMTS QoS CLASSES

3.9. UMTS QoS CLASSES With respect to Quality of Service, the UMTS is quite different from 2G or 2.5G network. The 2G network was designed to carry voice. Thus, the QoS was characterized by busy hour call blocking probability, call drop rate and voice quality. In GPRS (2.5G), even though primitive QoS features were introduced, the service remained a best effort. In contrast, the UMTS provides traffic with different bandwidth and QoS requirements. The QoS requirements mainly relate to QoS parameters like delay, delay variation and bit-error. Based on these QoS parameters, four different QoS classes are defined. These classes are as follows:

Conversational class

Streaming class

Interactive class

Background class

Table 3.2 provides a brief description of each QoS class and also the QoS requirements for each class. For example, Conversational Class has stringent requirements for delay and delay variation. This implies that for this class, delay and delay variation should be as low as possible so as to provide the desired service level. Each of the above QoS classes is further explained in the sections that follow.

Table 3.2 UMTS QoS Classes

QoS class Description Requirements on QoS parameters Applications

Delay Delay Bit-error Variation

Conversational Characterized by strict upper bounds on the transfer delay and Stringent Stringent No Voice and class delay variation. video conferencing

Streaming Characterized by real-time one-way data flow aimed at a Constrained Constrained No Video-on- class human user. demand

Interactive Characterized by a request-response protocol where a request Loose No Stringent Web- class is followed protocol where a request is followed browsing and Internet- based email

Background Characterized by no requirements on delay or delay variation No No Stringent E-mail, SMS class so that applications using it can run in back-ground. and FAX

For details of QoS classes, the reader is referred to 3GPP TS 23.107.

3.9.1. Conversational Class

Among all the QoS classes, the technical requirement of conversational class is most stringent. The stringency arises from strict upper bounds on transfer delay and delay variation. This implies that not only must the data streams reach the destination within a specified time period, the variation in time taken by different packets/streams to reach the destination must also be minimal. The bounds on delay and delay variation are governed by human perception of video and audio conversation. This QoS class, however, permits some data loss.

Given its conversational nature, the conversational class is also referred to as real-time service. Typical applications of conversational class include voice conversation and video conferencing.

3.9.2. Streaming Class

3.9.2. Streaming Class In terms of requirements for real-time transfer, this class comes after the conversational class. It is characterized by real-time data flow aimed at a human user. The flow is one-way and thus the name streaming class or streaming service.

Given the streaming and non-interactive nature of the application, the requirements for transfer delay are much less as compared to the conversational class. The delay variation requirement is similar to that of the conversational class, though somewhat less stringent. What this implies is that due to the non-interactive nature of the application, some delay is permitted in the time taken for the data stream to reach its destination. Further, with the use of buffering, minor variations in delay can also be tackled. However, if the delay variation exceeds a value beyond which buffering does not solve the problem, there is degradation in audio/video quality.

The streaming class is a relatively new concept. New 3G applications like video-on-demand belong to the streaming class.

3.9.3. Interactive Class

The interactive class is used by applications that adopt a request-response protocol where a request is followed by download of requested data. Given the request-response nature of those applications, there is an upper limit on the transfer delay. However, given the non real-time nature of communication, the requirement on the transfer delay is much less stringent as compared to the conversational class. Further, there is no requirement on delay variation. However, there are stringent requirements on data loss.

The applications using interactive class include Web-browsing, Internet-based email and data base queries.

3.9.4. Background Class

The background class is used by applications that do not have any delay or delay variation requirements. The only requirement is on data loss, which is not acceptable. This QoS class is named so because applications using it can run in the background (i.e. as a low priority task). This implies that given the real-time nature of other QoS classes, this class gets lower priority. Thus, this QoS class is served as a background activity as and when bandwidth is available in the network.

The applications using background class include email, SMS and FAX.

3.10. SUMMARY

This chapter provided an overview of the UMTS network architecture. The UMTS network is divided into three parts: UE, Access Network and Core Network. The next three chapters describe these three standard components.